unmatched tags in geocache reference file

[tsteven4]

1. gpsbabel/reference/gc/GCGCA8.gpx

   Line 46 in 8dae11e

   &lt;img SRC="http://www.mtgc.org/mtgc_member-banner.gif" WIDTH="500" HEIGHT="40" ALT="Member of Middle Tennessee GeoCachers Club [www.mtgc.org]" BORDER="0"&gt;&lt;/a&gt;&lt;/p&gt;</groundspeak:long_description>

   
   The tags </a></p> are unmatched.
   This causes the output from gpsbabel -i gpx -f reference/gc/GCGCA8.gpx -o html -F GCGCA8.html to contain unmatched tags.
2. There is an additional problem. The P end tag we add after gpsbabeldesclong is unmatched because of the UL tags in what we think is the paragraph content.

[robertlipe]

That HTML was created by hand (mine) that was rendered inside an iframe of browsers and using crudest tools almost 20 years ago. It's even possible that it wasn't even really written in 'real' HTML as I think the cache page editor of that time used sort of a bbcode markup that was transcribed on the server. As long as we can approximately round-trip what's inside groundspeak:long_description, what we do with that HTML is up to us. That cache placement has long since been removed and out of circulation. There's no need to overthink keeping it in sync with a master copy or whatever. We're free to change that reference file and the ones that come from it as we see fit. We're free to edit the past if it makes the present or future better. Signed, Winston Smith

…

On Sat, Jul 30, 2022 at 3:38 PM tsteven4 ***@***.***> wrote: 1. https://github.com/GPSBabel/gpsbabel/blob/8dae11e100677217c6b4cd858a76f3bf6bc8d12c/reference/gc/GCGCA8.gpx#L46 The tags </a></p> are unmatched. This causes the output from gpsbabel -i gpx -f reference/gc/GCGCA8.gpx -o html -F GCGCA8.html to contain unmatched tags. 2. There is an additional problem. The P end tag we add after gpsbabeldesclong is unmatched because of the UL tags in what we think is the paragraph content. — Reply to this email directly, view it on GitHub <#895>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACCSD34I3T52QZHZA4MA4ELVWWHEDANCNFSM55DYMFZA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[robertlipe]

Is there remaining action here? That embedded HTML snippet is from ancient browser days, when

s were considered self closing. (This was before even XHTML.) It's certainly in "quirks mode" even if it's not outright broken.

If you're OK with me deleting </a></p&gt at the end of long description and fixing any fallout reference files I'll do that. It's trying to close a link that's not here and a paragraph that was never opened. That's just copy-paste error in the source material from the Mozilla 1.5 era or so...

We're not testing the validity of our embedded HTML, so I don't think we have to faithfully reproduce/handle/fix what was clearly broken HTML.

I'm happy to do the fix if you agree that's OK.

[robertlipe]

Sticking a note to #1196

[tsteven4]

2. There is an additional problem. The P end tag we add after gpsbabeldesclong is unmatched because of the UL tags in what we think is the paragraph content.

Point 2. was fixed in dcdbb6a

[tsteven4]

1. The tags </a></p> are unmatched.

Point 1. is fixed in #1196, and updated version of GCGCA8.gpx.